Darshika Khandelwal
Python | Gen AI | SQL | Pyspark | Databricks
- Role
- Analyst at Deloitte
- Location
- Gurugram, HR, IN
- LinkedIn followers
- 500 followers
About Darshika Khandelwal
Working as a Data Analyst at Deloitte.
Experience
Analyst
Oct 2024 — Present · Gurgaon, IN
Worked on end-to-end forecasting solutions across E-Commerce and Organised Trade (OT) channels using sales, redistribution, campaign, inventory and macro datasets. Built and optimised forecasting pipelines using Microsoft Azure Data bricks for large scale data transformation and analytics workflows. Applied statistical and ML forecasting techniques including Croston, SES, DES, TES and Random Forest for demand prediction and model optimisation. Performed operational validations, data quality checks, model accuracy evaluation, and result collation to support enterprise scale forecasting ecosystems. Currently implementing the LOESS model to improve accuracy and automating the forecasting validation process to enhance operational efficiency, and data reliability Analysed large scale financial/logistics data using PySpark and SQL, building scalable pipelines for reporting and forecasting. Performed time-series and variance analysis (YoY, MoM, MTD) to derive regional revenue insights (AMEA, EU, AMER). Built analytical frameworks (Bridge-to-plan, GAF) for root cause analysis and forecasting accuracy. Designed a Generative AI prompt library to improve team productivity and standardised LLM usage. Developed use case specific prompt templates for tasks like data analysis, SQL generation, and documentation. Improve response quality and consistency by incorporating prompt structuring, context engineering, and iterative tuning. Enabled faster adoption of LLMs across the team by creating reusable workflows. Built a time series transaction dataset for customers to analyse their purchase behaviour. Developed a rule based approach to determine credit limits for existing customers and worked on a machine learning model to determine credit limits for new customers. This application helped users to generate detailed feature files from sales and collections data, and apply predictive models to estimate optimal credit limits, which helped in business growth.
Education
Ansal University Gurgaon
BTech cse, AI ML
2020 — 2024
Find verified contacts for anyone on LinkedIn
Unifers gives sales teams verified emails and direct dials, enriched profiles, and outreach that lands in the inbox.
Free plan included · No credit card required
This profile is compiled from publicly available professional sources. Unifers is not affiliated with or endorsed by LinkedIn. Request removal of this profile.